Content uploaded by Bertrand Brassart
Author content
All content in this area was uploaded by Bertrand Brassart on Sep 02, 2016
Content may be subject to copyright.
The key scientific and
medical informations
www.jle.com
Find the contents of this issue hereafter:
http://www.john-libbey-eurotext.fr/fr/
revues/medecine/ejd/sommaire.md?type=
text.html
Montrouge, 27-08-2016
Sylvie Brassart-Pasco
You will find below the electronic reprint of your article (pdf format):
Aging decreases collagen IV expression in vivo in the dermo-epidermal junction and in vitro in
dermal fibroblasts: possible involvement of TGF-ˇ1
published in:
European Journal of Dermatology, 2016, Volume 26, Num´
ero 4
John Libbey Eurotext
This electronic reprint is provided for your exclusive use, and can’t be communicated but for scientific and individual purposes. It mustn’t,
under any circumstances, be used or distributed for advertising, promotional or commercial purposes.
© John Libbey Eurotext, 2016
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
doi:10.1684/ejd.2016.2782
350 EJD, vol. 26, n◦4, July-August 2016
To cite this article: Feru J, Delobbe E, Ramont L, Brassart B, Terryn C, Dupont-Deshorgue A, Garbar C, Monboisse JC, Maquart FX, Brassart-Pasco S. Aging decreases
collagen IV expression in vivo in the dermo-epidermal junction and in vitro in dermal fibroblasts: possible involvement of TGF-1. Eur J Dermatol 2016; 26(4): 350-60
doi:10.1684/ejd.2016.2782
Investigative report Eur J Dermatol 2016; 26(4): 350-60
Jezabel FERU1
Etienne DELOBBE1
Laurent RAMONT1,2
Bertrand BRASSART1
Christine TERRYN3
Aurelie DUPONT-DESHORGUE1
Christian GARBAR4
Jean-Claude MONBOISSE1,2
Francois-Xavier MAQUART1,2
Sylvie BRASSART-PASCO1
1Laboratoire de Biochimie Médicale
et de Biologie Moléculaire,
CNRS UMR 7369: Matrice Extracellulaire
et Dynamique Cellulaire (MEDyC),
UFR Médecine, Université de Reims
Champagne-Ardenne,
51 rue Cognacq Jay, CS 30018, 51095
2Laboratoire Central de Biochimie,
Hôpital Robert Debré,
CHU de Reims,
Avenue du Général Koenig, 51092
3Plate-forme d’Imagerie Cellulaire
et Tissulaire IBISA,
Université de Reims Champagne-Ardenne,
51 rue Cognacq Jay, CS 30018, 51095
4Départements de Biopathologies,
Institut Jean-Godinot-Unicancer,
1 rue du général KOENIG CS80014 51726
Reims Cedex,
France
Reprints: S. Brassart-Pasco
<sylvie.brassart-pasco@univ-reims.fr>
Article accepted on 06/3/2016
Aging decreases collagen IV expression
in vivo in the dermo-epidermal junction
and in vitro in dermal fibroblasts: possible
involvement of TGF-1
Background: Collagen IV is a major component of the dermo-epidermal
junction (DEJ). Objectives: To study expression of collagen IV upon
aging in the DEJ and dermal fibroblasts isolated from the same patients.
A model of senescent fibroblasts was developed in order to identify bio-
logical compounds that might restore the level of collagen IV. Materials
& methods: Skin fragments of women (30 to 70 years old) were col-
lected. Localisation of collagen IV expression in the DEJ was studied by
immunofluorescence. Fibroblast collagen IV expression was studied by
real-time PCR, ELISA, and western blotting. Premature senescence was
simulated by exposing fibroblasts to subcytotoxic H2O2concentrations.
Results: Collagen IV decreased in the DEJ and fibroblasts relative to
age. TGF-1 treatment significantly increased collagen IV gene and
protein expression in fibroblasts and restored expression in the model
of senescence. Addition of TGF-1-neutralizing antibody to fibroblast
cultures decreased collagen IV expression. Conclusion: Taken together,
the results suggest that the decrease in collagen IV in the DEJ, relative to
age, could be due to a decrease in collagen IV expression by senescent
dermal fibroblasts and may involve TGF-1 signalling.
Key words: collagen IV, dermo-epidermal junction, fibroblast, stress-
induced premature senescence, skin, TGF-1
Skin aging is a complex biological phenomenon
consisting of two components: intrinsic aging and
extrinsic aging caused by environmental expo-
sure. Reactive oxygen species (ROS) are highly reactive
molecules that consist of a number of diverse chemical
species, including superoxide anion (O2-), hydroxyl rad-
ical (OH), and hydrogen peroxide (H2O2). Because of
their potential to cause oxidative deterioration of DNA,
proteins, and lipids, ROS have been implicated as one of
the causative factors of aging [1, 2]. Fibroblast premature
senescence can be triggered upon treatment with subtoxic
H2O2doses [3]. During cellular senescence, normal human
fibroblasts change their morphology from a spindle shape to
an enlarged, flattened and irregular shape [4]. An increase in
lysosomal -galactosidase activity with cellular senescence
was also reported and this senescence-associated -
galactosidase (SA--Gal) is considered as a marker of cel-
lular senescence and aging [5-7]. Cellular senescence is also
characterized by a reduced proliferation rate, an increased
number of cells in G0-G1 phase, and an increase in
p21Waf-1 expression [8]. In addition, changes in extracellu-
lar matrix macromolecule synthesis have also been reported
in senescent fibroblasts, and H2O2is reported to increase
the level of MMP-1 mRNA in human dermal fibroblasts
[9]. Finally, cellular senescence is also reported to decrease
type I collagen expression in dermal fibroblasts [10].
The TGF-signalling pathway is involved in the synthe-
sis of many extracellular matrix macromolecules in dermal
fibroblasts [11] and stimulates type I collagen biosynthesis
[12]. Previous studies from Mori et al. [13] demonstrated
that collagen I expression was decreased in in vitro-aged
fibroblast cultures. The mRNA levels of ␣1(I) collagen,
TGF-1 and TGF-receptor II (TGFRII) were reduced
to 62%, 62% and 59%, respectively, in late-passage fibrob-
lasts, compared to early-passage fibroblasts. Quan et al.
also demonstrated that the TGF-/Smad pathway is sig-
nificantly reduced in aged dermal fibroblasts in vivo and
mediates reduced collagen I expression [14]. TGF-1is
also reported to stimulate type IV collagen synthesis in
murine mesangial cells and mouse embryonic fibroblasts
[15, 16].
Basement membranes are thin and amorphous special-
ized extracellular matrices that play roles in various
biological events, including embryonic development. The
skin basement membrane binds the epidermis tightly to
the dermis, determines the polarity of the basal ker-
atinocytes, and acts as a selective barrier to control
molecular and cellular exchange. Type IV collagen is a
major component of skin basement membrane. Type IV
collagen molecules are heterotrimers composed of three
␣-chains, which exist in six genetically distinct forms:
␣1(IV) to ␣6(IV). COL4A1/COL4A2,COL4A3/COL4A4
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
EJD, vol. 26, n◦4, July-August 2016 351
and COL4A5/COL4A6 genes are located pairwise in
a head-to-head fashion on chromosomes 2, 3 and X,
respectively [17]. Each ␣chain is characterised by a long
collagenous domain of 1,400 residues of Gly-X-Y repeats,
interrupted by a short non-collagenous sequence, and a
non-collagenous (NC1) domain of 230 residues at the car-
boxyl terminus, which is conserved among different species
[18]. Only two collagen IV isoforms have been reported
in the adult skin basement membrane: ␣1(IV)2/␣2(IV), the
major isoform; and ␣5(IV)2/␣6(IV), the minor isoform [19-
21]. Temporal and spatial expression of ␣1(IV), ␣2(IV),
␣5(IV) and ␣6(IV) collagen chains was studied during the
formation of the basal lamina in an in vitro skin model
[22]. A sequential study was performed with 7-day and
14-day cultures (lamina densa absent), and with 28-, 36-,
and 56-day cultures (lamina densa present). In the pres-
ence of the lamina densa, type IV collagen chains were
mainly produced in the dermis. Type IV collagen expres-
sion by dermal fibroblasts was also reported by Sasaki
et al. [23]. A study from Vazquez and collaborators [24]
demonstrated decreased type IV collagen in skin samples
taken from a site, 6 cm above the pubic symphysis, in
women aged 35 to 60. In 1998, Le Varlet and collab-
orators [25] also studied the influence of donor age on
collagen expression in post-auricular skin biopsies from
four donors (women aged 10, 35, 50, and 64, respectively)
and showed less type IV collagen in adult dermo-epidermal
junctions, however, the cause of this decrease is currently
unknown.
In the present study, we set out to investigate whether
type IV collagen, as well as TGF-1 and TGFRII, is
decreased in the skin basement membrane upon aging.
As fibroblast premature senescence can be triggered by
treatment with subtoxic H2O2doses [3], we further inves-
tigated type IV collagen expression in this model of
stress-induced premature senescence (SIPS). The effect of
TGF-1 on type IV collagen expression was also inves-
tigated in dermal fibroblast cultures, as well as the SIPS
model.
Materials and methods
Reagents
Culture media and trypan blue were purchased from Invit-
rogen (distributed by Fischer Scientific, Illkirch, France),
TGF-1 and Senescence Cells Histochemical Staining
Kit from Sigma Aldrich (Saint-Quentin Fallavier, France),
Qiagen RNeasy kit from Qiagen (Courtaboeuf, France),
Maxima First Strand cDNA synthesis kit from Fermen-
tas France/Thermo-Fisher Scientific (Villebon sur Yvette,
France), SYBR Premix Ex Taq kit from Ozyme (Saint
Quentin en Yvelines, France), and anti-type IV collagen
from Dako (Les Ulis, France).
Ethics statement
Collection and utilisation of human skin fragments were
approved by the Institutional Review Board of the Reims
University Hospital (CHU de Reims), and conducted
according to the declaration of Helsinki principles. A writ-
ten informed consent was obtained from patients.
Cell culture
Skin fragments from patients from 30 to 70 years old,
following breast surgery, were cleaned of excess subcuta-
neous tissue and cut into small pieces (5×5 mm), and then
incubated with 0.25% trypsin in 0.01% ethylenediamine
tetraacetic acid solution at +4 ◦C overnight. The epidermal
sheets were separated from the dermis by forceps. Primary
cultures of dermal fibroblasts were established by cutting
the dermis into small pieces (1×1 mm) and placing the frag-
ments into 25 cm2flasks. Cells were cultured in Dulbecco’s
modified Eagle’s medium supplemented with 10% foetal
bovine serum (FBS) and incubated at +37 ◦C in a humidi-
fied atmosphere containing 5% CO2. Fibroblasts were then
trypsinized. Cultures between passages 3 and 5 were used
in this study.
Induction of fibroblast senescence
by H2O2treatment
Fibroblasts were seeded in 6-well plates and grown to 80%
confluence. The culture media were removed and the cells
were exposed to fresh culture medium containing 0 to
100 MH
2O2. After 1 hour and 30 minutes of H2O2expo-
sure, the culture medium was removed and the cells were
incubated with fresh culture medium for 48 hours.
Trypan blue exclusion test
The effect of H2O2exposure on cell viability was deter-
mined using the trypan blue exclusion test. Cells were
treated in the absence or presence of various concentra-
tions of H2O2for 48 hours. Then, cells were trypsinized and
diluted in phosphate-buffered saline. The floating dead cells
in the medium and cells that remained attached to the plates
were then counted using an automated cell counter (Count-
ess, Invitrogen, distributed by Fischer Scientific, Illkirch,
France) in the presence of trypan blue solution at a 1:1 ratio
(v/v) (Sigma), as described by the manufacturer.
Detection of senescence-associated
-galactosidase (SA--Gal) activity
Forty-eight hours after H2O2exposure, the culture media
were removed and the cells were stained using the Senes-
cence Cells Histochemical Staining Kit (Sigma-Aldrich,
Saint-Quentin Fallavier, France).
Quantitative real-time PCR
RNA isolation was performed using the Qiagen RNeasy kit
(Qiagen, Courtaboeuf, France), according to the manufac-
turer’s instructions. cDNA was prepared from 1 g of total
RNA by reverse transcription (RT) at +42 ◦C for 45 min
using the Maxima First Strand cDNA synthesis kit (Fermen-
tas France/Thermo-Fisher Scientific, Villebon sur Yvette,
France). SYBR Premix Ex Taq (Ozyme, Saint Quentin en
Yvelines) was used for the PCR reaction. Primer sequences
are listed in table 1. Mx 3005P (Stratagene) was used for
amplification and data collection. A first denaturation step
was performed for 10 minutes at +95 ◦C, then a cycling pro-
gram (40 cycles) included a 5-minute denaturation step at
+95 ◦C, a 30-second annealing step at +60 ◦C, followed by
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
352 EJD, vol. 26, n◦4, July-August 2016
Table 1. List of the primer sequences used in real-time PCR experiments.
Gene Forward primer (5’→3’) Reverse primer (5’→3’)
Human ␣1(IV) TAGAGAGGAGCGAGATGTTC GTGACATTAGCTGAGTCAGG
Human GAPDH ACGGATTTGGTCGTATTGGG TGATTTTGGAGGGATCTCGC
Human MMP-1 TGGCATTCAGTCCCTCTATGG AGGACAAAGCAGGATCACAGTT
Human p21WAF-1 CTGGAGACTCTCAGGGTCGAA CCAGGACTGCAGGCTTCCT
Human ␣1(I) CACCAATCACCTGCGGTACAGAA CAGATCACGTCATCGCACAAC
Human TGF-1 GCGTGCTAATGGTGGAAAC CGGTGACATCAAAAGATAACCAC
Human TGFRII GTCTACTCCATGGCTCTGGT ATCTGGATGCCCTGGTGGTT
a 15-second elongation step at +72 ◦C. Fluorescence acqui-
sition was carried out at +72 ◦C in single mode at the end
of the elongation step. After real-time PCR, melting curve
analysis was performed by continuously measuring fluores-
cence during heating from +55 to +95 ◦C at a transition rate
of +0.2◦C/s. Product specificity was evaluated by melting
curve analysis and electrophoresis on 2% agarose gels. Flu-
orescence was analysed using the Data Analysis software
(Stratagene). Crossing points (Cp or Ct) were established
using the second derivative method. Real-time PCR effi-
ciency was calculated from the slope of the standard curve.
Target gene expression levels were normalised to reference
genes. The results were calculated using the delta-delta
method.
Immunohistochemistry
Breast samples were collected from 30 to 70-year-old
women. Skin sections, 5 m thick, were prepared using
a cryostat and placed on Superfrost slides. These slides
were then washed twice in distilled water for 5 minutes and
twice in PBS for 10 minutes. They were then incubated with
anti-collagen IV primary antibody (DAKO) (diluted 1:200)
for 45 minutes in a humid chamber at room temperature.
They were then rinsed in PBS for 10 minutes and incubated
for 10 minutes with the Alexa Fluor 488-conjugated sec-
ondary antibody (diluted at 1:200). They were rinsed with
PBS for 10 minutes and then with distilled H2O for 10 min-
utes. Slides were examined under a confocal laser scanning
microscope (Zeiss LSM 710) (Carl Zeiss MicroImaging
GmbH, Germany). Two image processing programs based
on ImageJ application were developed to measure basement
membrane invaginations and basement membrane type IV
collagen.
Western blot
Conditioned media were harvested, centrifuged at 500 g
for 10 min at 4 ◦C to remove cellular debris, and concen-
trated using Sartorius Vivaspin concentrators (distributed
by Dominique Dutscher, Brumath, France). The protein
content of the media was determined by the Bradford
method, using bovine serum albumin as a standard. 50 g
of protein was separated by SDS-PAGE and transferred to
an Immobilon-P membrane (Millipore, St Quentin en Yve-
lines, France). After blocking with 5% non-fat dry milk in
TBS buffer (50 mM Tris, 138 mM NaCl, 2.7 mM KCl,
pH 8.0) containing 0.1% Tween-20 (TBS-T), the mem-
brane was incubated with an anti-type IV collagen rabbit
polyclonal antibody (1/5,000 in TBS-T containing 1% non-
fat dry milk) overnight at 4◦C. Then, the membrane was
washed in TBS-T buffer and incubated with a peroxidase-
conjugated goat anti-rabbit IgG (1/10,000) for 1 hour and
30 minutes. Signals were detected using an ECL+ kit (GE
Healthcare, Orsay, France), according to the manufacturer’s
instructions. Membranes were stained with Coomassie blue
and the total amount of protein per well was determined
using Image Lab software. Collagen IV was quantified rel-
ative to total loaded protein.
Enzyme-linked immunosorbent assay
Type IV collagen was determined using the ELISA Kit for
Collagen Type IV (Uscn Life Science Inc., Euromedex,
Souffelweyersheim, France), according to the manufacturer
instructions.
Statistical analyses
For in vitro experiments, the student t-test was performed
and results are expressed as mean ±SD. For in vivo exper-
iments, comparisons between the two groups were made
using the non-parametric u-test of Mann and Whitney and
results are depicted as box plots.
Results
Aging decreases basement membrane
type IV collagen in the skin
Transversal skin sections were prepared from breast frag-
ments derived from 30 to 70-year-old women. Localisation
of type IV collagen expression was studied by immunoflu-
orescence (figure 1A). Type IV collagen was localised to
blood vessels and the dermo-epidermal junction (DEJ).
Invagination of the basement membrane decreased rela-
tive to age (figure 1B). Image analysis of the skin sections
demonstrated that DEJ type IV collagen decreased by 36%
in the skin of women aged 50-70 versus 30-50 (figure 1C).
Aging decreases type IV collagen, TGF1
and TGFRII expression in dermal fibroblasts
Fibroblasts were isolated from breast skin fragments
derived from 30 to 70-year-old women. Type IV
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
EJD, vol. 26, n◦4, July-August 2016 353
B
A
C
1.5
1.0
0.5
0.0 31-40 51-60
***
***
***
41-50 61-70
Age (years)
Basement membrane
invagination (A.U.)
Type IV collagen
quantification (A.U.)
30-50 50-70
***
Age (years)
0
2
4
5
6
3
1
35 years 65 years
BM BM
Figure 1. Quantification of type IV collagen in skin basement membrane. A) Frozen skin sections were immunostained with
an anti-collagen IV primary antibody; scale bar: 1 mm. B) Quantification of basement membrane (BM) invaginations relative to
age. C) Quantification of type IV collagen in the DEJ using an image processing program based on ImageJ application.
For each age group, n= 16; ***p<0.001.
collagen gene expression was studied by real-time PCR;
␣1(IV) chain expression was decreased by 66% in
women aged 50-70 versus 30-50 (figure 2A) and this
was also the case for ␣2(IV) chain expression (data
not shown). ␣5(IV) and ␣6(IV) chains were expressed
at a much lower level compared to ␣1(IV) and ␣2(IV)
chains (data not shown). Type IV collagen protein
level was quantified by ELISA, and was shown to
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
354 EJD, vol. 26, n◦4, July-August 2016
CD
AB
α1(IV)/GAPDH (x10-2)
2.500
2.000
1.500
1.000
0.250
0.125
30-50 50-70
***
Age (years)
0.500
Collagen IV (pg/mL)
0.4
0.3
0.2
0.1
0.0
30-50 50-70
**
Age (years)
TGFβ1/GAPDH
0.3
0.2
0.0
30-50 50-70
**
Age (years)
0.1
TGFβRII/GAPDH
Age (years)
0.8
0.6
0.4
0.2
0.0
30-50 50-70
***
Figure 2. Quantification of type IV collagen expression in fibroblasts. Fibroblasts were isolated from breast fragments derived
from 30 to 70-year-old women. A) Type IV collagen gene expression was investigated by real-time PCR. B) Protein expression
was measured by ELISA. TGF-1(C) and TGFRII (D) gene expression analysed by real-time PCR. For each age group, n= 16;
**p<0.005; ***p<0.001.
decrease by 35% in women aged 50-70 versus 30-50
(figure 2B).
Since TGF-1 is reported to modulate type IV collagen and
␣1(I) collagen synthesis in mouse embryonic fibroblasts,
we investigated TGF1 and TGFRII gene expression
for both age groups. TGF1 and TGFRII mRNA levels
decreased by 68% and 70%, respectively, in women aged
50-70 versus 30-50 (figure 2D).
Induction of SIPS phenotype
in dermal fibroblasts
We then studied type IV collagen expression in a model
of accelerated senescence. Fibroblasts were treated for
48 hours with 0 to 200 MH
2O2and cell viability was
assessed using the trypan blue exclusion test (figure 3A).
Treatments with 50 and 100 MH
2O2did not alter
cell viability (98.8% and 97.3% viability, respectively).
A 200-MH
2O2treatment decreased cell viability
(84.6% viability). We decided to use 50 and 100 M
H2O2to induce cell senescence without affecting cell
viability. Fibroblast morphology was analysed using an
inverted microscope. After treatment with 50 MH
2O2,
cells appeared slightly flattened and this morphology
was exaggerated after treatment with 100 MH
2O2
(figure 3B). To confirm the senescent phenotype, SA--Gal
activity was measured. Treatment with 50 MH
2O2
triggered an increase in SA--Gal activity that was
slightly amplified after treatment with 100 MH
2O2
(figure 3C). Cellular senescence is also reported to be char-
acterised by a reduced proliferation rate and an increase
in p21Waf-1 expression. p21Waf-1 expression, analysed
by real-time PCR, was increased by 42.2 % and 78.1%
after treatments with 50 and 100 MH
2O2, respectively
(figure 3D). A decrease in type I collagen expression and
increase in MMP-1 expression were reported in senescent
fibroblasts. Based on real-time PCR analysis, we demon-
strated that 50-M and 100-MH
2O2treatments decreased
COL1A1 expression by 60.8% and 88.2%, respectively
(figure 3E), whereas MMP-1 expression, also determined
by real-time PCR, was increased by 42% and 73% in
the presence of 50 M and 100 MH
2O2treatments,
respectively (figure 3F). Collectively, these results confirm
that treatment with 100 MH
2O2induces a senescent
phenotype in fibroblasts without affecting cell viability.
Inhibition of collagen IV expression
in fibroblasts after H2O2treatment
Type IV collagen gene expression was studied in the
SIPS model by real-time PCR. COL4A1 gene expres-
sion was significantly decreased after treatment with
50 and 100 MH
2O2(-35.4% and -50.6%, respec-
tively) (figure 4A). COL4A2 gene expression was similarly
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
EJD, vol. 26, n◦4, July-August 2016 355
C
0 50 100
B
A120
100
Viabiliy (%)
100 150 200
80
60
50
40
20
0
0
0 50 100
[H2O2] (μM)
[H2O2] (μM)
[H2O2] (μM)
F
D
10050
0.14
0
p21WAF -/GAPDH
0.00
0.02
0.04
0.06
0.08
0.10
0.12 **
***
[H2O2] (μM) [H2O2] (μM)[H2O2] (μM)
E
10050
0.80
0
α1(I)/GAPDH
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
***
***
MMP-1/GAPDH
10050
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0
***
***
Figure 3. Validation of the fibroblast senescent phenotype after H2O2treatment. A) Cell viability assessed by trypan blue
exclusion. B) Cell morphology observed under an inverted microscope; scale bar: 10 m. C) SA--Gal activity observed under
an inverted microscope; senescent cells are stained blue; scale bar: 10 m. Real-time PCR of p21Waf-1 (D), COL1A1 (E), and
MMP-1 (F). Results are expressed as mean ±SD; **p<0.005; ***p<0.001.
decreased (data not shown). The level of type IV collagen
in the SIPS model was also analysed by western blot. The
level of type IV collagen was significantly decreased after
treatment with 50 and 100 MH
2O2(-30.5% and -49.1%,
respectively) (figure 4B).
Effect of exogenous TGF-1 and
TGF-1-neutralizing antibody on type IV
collagen in normal dermal fibroblasts
We tested the effect of TGF-1 on type IV colla-
gen expression in normal dermalfibroblasts. COL4A1
gene expression, analysed by real-time PCR, was
increased in response to TGF-1 in a dose-dependent
manner. The effect was significant above a threshold
of 0.5 ng/mL and increased progressively with a max-
imal stimulation at 10 ng/mL (figure 5A). Collagen IV,
analysed by western blotting, was significantly increased
above a threshold of 0.5 ng/mL and increased pro-
gressively with a maximal stimulation at 10 ng/mL
(figure 5B).
Normal dermal fibroblasts were incubated with or with-
out TGF-1-neutralizing antibody for 48 hours. The level
of collagen IV, analysed by western blotting, was signifi-
cantly decreased in the presence of the blocking antibody
(figure 6).
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
356 EJD, vol. 26, n◦4, July-August 2016
0.0
0.5
1.0
1.5
2.0
2.5
3.0
A
α1(IV)/GAPDH (x10
-2
)
**
**
500 100
[H2O2] (μM)
[H2O2] (μM)
B
500 100
Coomassie blue staining
[H2O2] (μM)
Collagen IV
50
0 100
Western blot
[H2O2] (μM)
0
***
**
0,0000
0,0010
Collagen IV/total protein
0,0015
0,0020
0,0025
0,0005
50 100
Figure 4. Type IV collagen expression in fibroblasts after H2O2treatment. Fibroblasts were incubated with 0, 50 or 100 M
H2O2for 1 hour and 30 minutes, washed, and further incubated for 48 hours. A)COL4A1 gene expression measured by real-time
PCR. B) Type IV collagen protein analysed by western blotting and quantified; the membrane was stained with Coomassie blue
to control for loaded protein. Results are expressed as mean ±SD; **p<0.005; ***p<0.001.
Effect of TGF-1 on type IV collagen
expression in senescent fibroblasts
Fibroblasts were treated with 100 MH
2O2for 1 hour and
30 minutes. The medium was then removed and cells were
incubated without H2O2for 48 hours until induction of
senescence. Type IV collagen gene and protein levels were
shown to decrease after 48 hours (figure 7A, B). Senes-
cent cells were further treated with or without TGF-1
(10 ng/mL). TGF-1 treatment completely restored type
IV collagen gene and protein expression in the senescent
fibroblasts (figure 7C, D).
Discussion
During aging, the extracellular matrix undergoes significant
alterations. Using immunohistochemistry and transmission
electron microscopy, Vazquez et al showed that type IV
collagen was decreased with aging in skin samples, taken
from a site 6 cm above the pubic symphysis of women
from 35 to 60 years old [24]. Le Varlet et al. also stud-
ied the influence of donor age on collagen expression in
post-auricular skin biopsies from four donors (women aged
10, 35, 50, and 64, respectively). Post-embedding type IV
collagen immunostaining and image analysis showed less
type IV collagen in adult dermo-epidermal junctions [25].
In the current study, we confirmed the decrease in type
IV collagen in the skin basement membrane of women
aged 50-70 compared to women aged 30-50. As type IV
collagen from the dermo-epidermal junction is mainly pro-
duced by fibroblasts [22, 23], we studied type IV collagen
expression in fibroblasts derived from women aged 30-70
years. ␣1(IV) chain gene expression decreased in women
aged 50-70, compared to women aged 30-50, and a sim-
ilar decrease in ␣2(IV) chain gene expression was also
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
EJD, vol. 26, n◦4, July-August 2016 357
A
TGF-β1 (ng/mL)
0.50
0
5
10
15
20
25
30
35
40
α1(IV)/GAPDH X10-2
1 2.5 5 10
***
***
***
**
BCoomassie blue staining
TGF-β1 (ng/mL)
0 0,5 1 2,5 5 100
Collagen IV
Western blot
TGF-β1 (ng/mL)
0,5 1 2,5 5 10
0.0014
0.0012
TGF-β1 (ng/mL)
Collagen IV/total protein
0.50
*
**
***
***
***
1 2.5 5 10
0.0010
0.0008
0.0006
0.0004
0.0002
0.0000
Figure 5. Stimulation of type IV collagen expression by TGF-1 in normal fibroblasts. Fibroblasts were stimulated with increasing
concentrations of TGF-1 for 48 hours. A)COL4A1 gene expression measured by real-time PCR. B) Protein analysed by western
blotting and quantified; the membrane was stained with Coomassie blue to control for loaded protein. Results are expressed as
mean ±SD; *p<0.01; **p<0.005; ***p<0.001.
observed (data not shown). This result is not surprising,
as both genes are under the control of a bidirectional pro-
moter [17]. ␣5(IV) and ␣6(IV) chains were expressed at a
much lower level, compared to ␣1(IV) and ␣2(IV) chains
(data not shown). This may explain the skin basement
membrane type IV collagen decrease with aging. Previ-
ous studies from Mori et al. [13] reported a decrease in
␣1(I) collagen, TGF-1, and TGF-receptor II (62%,
62% and 59%, respectively) in late-passage, compared
to early-passage, fibroblasts, suggesting that TGF-sig-
nalling plays a critical role in collagen I regulation. We
have confirmed the decrease in TGF-1 and TGF-RII in
aged-fibroblasts.
Since it appears to be difficult to study type IV collagen
regulation relative to aging in fibroblasts from patients due
to inter-individual variation, we decided to study type IV
expression in a model of accelerated fibroblast senescence.
We developed a stress-induced model of cell senescence
based on H2O2treatment. The senescent phenotype was
assessed using different methods. During cellular senes-
cence, normal human fibroblasts change their morphology
from a spindle shape to an enlarged, flattened and irregular
shape [4]. These alterations were clearly observed in our
model. In addition, an increase in SA--Gal activity was
reported [5-7], which we also observed in our model.
Cellular senescence is also characterised by a reduced rate
of proliferation, an increased number of cells in G0-G1, and
consequently an increase in p21Waf-1 expression [8]. We
observed an increase in p21Waf-1 expression in our model.
Collectively, our results confirm the senescent phenotype of
fibroblasts treated with 100 MH
2O2. A change in matrix
macromolecule expression was also reported in stress-
induced fibroblasts, such as a decrease in type I collagen
expression [10] and an increase in MMP-1 expression [9].
Our results are consistent with those previously reported
and validate our aging model. We found that type IV
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
358 EJD, vol. 26, n◦4, July-August 2016
Control
TGF-β1 neutr. Ab
Irrelevent Ab
Coomassie blue stainingWestern blot
Collagen IV
Control
TGF-β1 neutr. Ab
Irrelevent Ab
0.0030
***
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
Control
Collagen IV/total protein
TGF-β1 neutr. Ab
Irrelevent Ab
Figure 6. Decrease in type IV collagen expression in normal fibroblasts in the presence of TGF-1-neutralizing antibody.
Fibroblasts were incubated without antibody (control), with 1 g/mL TGF-1-neutralizing antibody (TGF-1-neutr. Ab) or
with 1 g/mL IgG1 (irrelevant Ab) for 48 hours. The level of collagen IV was analysed by western blotting and quantified; the
membrane was stained with Coomassie blue to control for loaded protein. Results are expressed as mean ±SD; ***p<0.001.
collagen expression was largely decreased after
H2O2treatment. This result suggests that the decrease in
basement membrane type IV collagen expression in the
skin could be due, at least in part, to a decrease in type
IV collagen expression in senescent dermal fibroblasts.
It was previously reported that such senescent cells exist
and accumulate with age in vivo [26]. We then tested the
effect of TGF-1 on type IV collagen expression in dermal
fibroblasts and found that TGF-1 largely increased type
IV collagen expression. This is in accordance with previous
results which showed that TGF-1 stimulates type IV col-
lagen expression in murine mesangial cells [15] and mouse
embryo NIH-3T3 fibroblasts [16]. Moreover, incubation
of normal dermal fibroblasts with TGF-1-neutralizing
antibody led to a decrease in collagen IV expression and
this highlights the involvement of TGF-1 in the control
of type IV collagen levels. We tested the effect of TGF-1
in the SIPS model and found that TGF-1 restored type
IV collagen expression under these conditions. The level
of TGF-1 was reported to decrease during fibroblast
replicative senescence [27]. Taken together, the results
suggest that type IV collagen decrease with aging involves
a TGFsignalling pathway.
Future research, following on from this study, should aim to
elucidate the mechanisms leading to collagen IV decrease at
the dermo-epidermal junction level with a view to propos-
ing strategies to restore expression, and at the same time,
maintain the integrity of the dermo-epidermal basement
membrane during aging. The SIPS model of dermal fibrob-
lasts thus represents a simple model to study type IV
collagen regulation over a relatively short period of time
in response to bioactive compounds for cosmetic purposes.
A similar decrease in collagen IV expression was obtained
in a replicative senescence model, based on cell cultures
from different passages (figure 8), however, this model is
less suited to rapid screening.
Non-collagenous NC1(␣1[IV]) (arresten) and
NC1(␣2[IV]) (canstatin) domains have anti-angiogenic
and/or anti-tumour properties [28]. The reduced level of
type IV collagen expression in the DEJ could therefore
play a role in the development of skin cancer in the elderly.
Further elucidation of these molecular mechanisms may
provide clues for future oncologic treatment.
Since type IV collagen chains are reported to be produced
largely by fibroblasts in an in vitro skin model [15], we
focused on type IV expression in fibroblasts during phys-
iological aging. Nevertheless, type IV collagen is also
expressed in keratinocytes and preliminary studies per-
formed in our laboratory demonstrate that COL4A1 and
COL4A2 gene expression is also decreased in aged ker-
atinocytes, even though these cells appear to produce five
times less type IV collagen relative to fibroblasts (data not
shown).
Disclosure. Financial support: This work was sup-
ported by the CNRS (PIR Longévité et vieillissement), the
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
EJD, vol. 26, n◦4, July-August 2016 359
0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
100
t = 48h
**
H2O2 (μM)
H2O2 (μM) H2O2 (μM)
α1(IV)/GAPDH (x10-2)
A
0
3.0
2.5
2.0
1.5
1.0
0.5
0.0
100
TGF-β1 (ng/mL)
α1(IV)/GAPDH (x10-2)
t = 96h
***
C
Coomassie
blue staining
010
TGF-β1 (ng/mL)
D
10
Western blot
0
TGF-β1 (ng/mL)
Collagen IV
Collagen IV/total protein
0.0030
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
***
Coomassie
blue staining
0100
B
100
Western blot
Collagen IV
0
Collagen IV/total protein
***
0.0030
0.0025
0.0020
0.0015
0.0010
0.0005
0.0000
Figure 7. Restoration of type IV collagen expression by TGF-1 in senescent fibroblasts. Fibroblasts were incubated with
100 MH
2O2for 1 hour and 30 minutes, washed, and further incubated for 48 hours. A)COL4A1 gene expression at 48 hours
measured by real-time PCR. B) Protein analysed by western blotting and quantified. At the end of the first 48-hour incubation
period, the senescent fibroblasts were then treated with 10 ng/mL TGF-1 for another 48-hour incubation period. C)COL4A1
gene expression was measured after the 48-hour incubation period (total incubation time: 96 hours). D) Protein was analysed by
western blotting after a 48-hour incubation period (total incubation time: 96 hours) and quantified; the membranes were stained
with Coomassie blue to control for loaded protein. Results are expressed as mean ±SD; **p<0.005; ***p<0.001.
0.10
α1(IV)/GAPDH
0.08
0.06
0.04
0.02
0.00
410
Number of passage
0.12
α2(IV)/GAPDH
0.10
0.08
0.06
0.04
0.02
0.00 410
***
***
Number of passage
Figure 8. ␣1 (IV) and ␣2 (IV) gene expressions were studied by real time PCR in 4-passage and 10-passage fibroblat cultures.
*** p<0.001.
Author offprint
© John Libbey Eurotext, 2016
Journal Identification = EJD Article Identification = 2782 Date: August 2, 2016 Time: 2:57pm
360 EJD, vol. 26, n◦4, July-August 2016
University of Reims Champagne-Ardenne. Jezabel FERU
was a PhD student with a fellowship from the Région
Champagne-Ardenne. Conflict of interest: none.
References
1. Harman D. The free radical theory of aging. Antioxid Redox Signal
2003; 5: 557-61.
2. Long YC, Tan TM, Takao I, Tang BL. The biochemistry and cell biol-
ogy of aging: metabolic regulation through mitochondrial signaling.
Am J Physiol Endocrinol Metab 2014; 306: E581-91.
3. Bladier C, Wolvetang EJ, Hutchinson P, de Haan JB, Kola I.
Response of a primary human fibroblast cell line to H2O2: senescence-
like growth arrest or apoptosis? Cell Growth Differ 1997; 8: 589-98.
4. Chen QM, Tu VC, Catania J, Burton M, Toussaint O, Dilley T.
Involvement of Rb family proteins, focal adhesion proteins and protein
synthesis in senescent morphogenesis induced by hydrogen peroxide.
J Cell Sci 2000; 113: 4087-97.
5. Lee BY, Han JA, Im JS, et al. Senescence-associated beta-
galactosidase is lysosomal beta-galactosidase. Aging Cell
2006; 5: 187-95.
6. Maier AB, Westendorp RG, Van Heemst D. Beta-galactosidase
activity as a biomarker of replicative senescence during the course
of human fibroblast cultures. Ann N Y Acad Sci 2007; 1100: 323-32.
7. Itahana K, Campisi J, Dimri GP. Methods to detect biomarkers of cel-
lular senescence: the senescence-associated beta-galactosidase assay.
Methods Mol Biol 2007; 371: 21-31.
8. Chen QM. Replicative senescence and oxidant-induced premature
senescence. Beyond the control of cell cycle checkpoints. Ann N Y
Acad Sci 2000; 908: 111-25.
9. Brenneisen P, Briviba K, Wlaschek M, Wenk J, Scharffetter-
Kochanek K. Hydrogen peroxide (H2O2) increases the steady-state
mRNA levels of collagenase/MMP-1 in human dermal fibroblasts. Free
Radic Biol Med 1997; 22: 515-24.
10. Bigot N, Beauchef G, Hervieu M, et al. NF-B accumulation
associated with COL1A1 transactivators defects during chronological
aging represses type I collagen expression through a -112/-61-bp
region of the COL1A1 promoter in human skin fibroblasts. J Invest
Dermatol 2012; 132: 2360-7.
11. Rolfe KJ, Irvine LM, Grobbelaar AO, Linge C. Differential gene
expression in response to transforming growth factor-beta1 by fetal
and postnatal dermal fibroblasts. Wound Repair Regen 2007;15: 897-
906.
12. Chen SJ, Yuan W, Mori Y, Levenson A, Trojanowska M, Varga J.
Stimulation of type I collagen transcription in human skin fibroblasts by
TGF-beta: involvement of Smad 3. J Invest Dermatol 1999; 112: 49-57.
13. Mori Y, Hatamochi A, Arakawa M, Ueki H. Reduced expression
of mRNA for transforming growth factor beta (TGF beta) and TGF beta
receptors I and II and decreased TGF beta binding to the receptors in
in vitro-aged fibroblasts. Arch Dermatol Res 1998; 290: 158-62.
14. Quan T, Shao Y, He T, Voorhees JJ, Fisher GJ. Reduced expression
of connective tissue growth factor (CTGF/CCN2) mediates colla-
gen loss in chronologically aged human skin. J Invest Dermatol
2010; 130: 415-24.
15. Silbiger S, Lei J, Ziyadeh FN, Neugarten J. Estradiol reverses
TGF-beta1-stimulated type IV collagen gene transcription in murine
mesangial cells. Am J Physiol 1998; 274: 1113-8.
16. Grande J, Melder D, Zinsmeister A, Killen P. Transforming growth
factor-beta 1 induces collagen IV gene expression in NIH-3T3 cells.
Lab Invest 1993; 69: 387-95.
17. Hudson BG, Reeders ST, Tryggvason K. Type IV collagen: struc-
ture, gene organization, and role in human diseases. Molecular basis
of Goodpasture and Alport syndromes and diffuse leiomyomatosis.J
Biol Chem 1993; 268: 26033-6.
18. LeBleu VS, Macdonald B, Kalluri R. Structure and function of base-
ment membranes. Exp Biol Med (Maywood) 2007; 232: 1121-9.
19. Tanaka K, Iyama K, Kitaoka M, et al. Differential expression
of alpha 1(IV), alpha 2(IV), alpha 5(IV) and alpha 6(IV) collagen
chains in the basement membrane of basal cell carcinoma. Histochem
J1997; 29: 563-70.
20. Tanaka N, Tajima S, Ishibashi A, et al. Expression of the alpha1-
alpha6 collagen IV chains in the dermoepidermal junction during
human foetal skin development: temporal and spatial expression of
the alpha4 collagen IV chain in an early stage of development.BrJ
Dermatol 1997; 139: 371-4.
21. Hasegawa H, Naito I, Nakano K, et al. The distributions of type
IV collagen alpha chains in basement membranes of human epidermis
and skin appendages. Arch Histol Cytol 2007; 70: 255-65.
22. Fleischmajer R, Kuhn K, Sato Y, et al. There is temporal and spatial
expression of alpha1 (IV), alpha2 (IV), alpha5 (IV), alpha6 (IV) colla-
gen chains and beta 1 integrins during the development of the basal
lamina in an “in vitro” skin model. J Invest Dermatol 1997; 109: 527-
33.
23. Sasaki S, Zhou B, Fan WW, et al. Expression of mRNA for type
IV collagen alpha1, alpha5 and alpha6 chains by cultured dermal
fibroblasts from patients with X-linked Alport syndrome. Matrix Biol
1998; 17: 279-91.
24. Vázquez F, Palacios S, Alema˜
n N, Guerrero F. Changes of the
basement membrane and type IV collagen in human skin during aging.
Maturitas 1996; 25: 209-15.
25. Le Varlet B, Chaudagne C, Saunois A, et al. Age-related functional
and structural changes in human dermo-epidermal junction compo-
nents. J Investig Dermatol Symp Proc 1998; 3: 172-9.
26. Dimri GP, Lee X, Basile G, et al. A biomarker that identifies senes-
cent human cells in culture and in aging skin in vivo. Proc Natl Acad
Sci USA 1995; 92: 9363-7.
27. Zeng G, McCue HM, Mastrangelo L, Millis AJ. Endogenous
TGF-beta activity is modified during cellular aging: effects on metal-
loproteinase and TIMP-1 expression. Exp Cell Res 1996; 228: 271-6.
28. Monboisse JC, Oudart JB, Ramont L, Brassart-Pasco S, Maquart
FX. Matrikines from basement membrane collagens: a new anti-cancer
strategy. Biochim Biophys Acta 2014; 1840: 2589-98.
Author offprint
